Supplementary Materials: WEF System Integration in Karst Regions – Model Data and Appendices

Description This repository contains the supplementary materials supporting the MILP-based techno-economic optimization of water-energy-food (WEF) system integration in karst regions, using Bijie City, Guizhou Province, Southwest China, as the case study. The research addresses a structural paradox common to karst landscapes: seasonal hydropower curtailment during wet months coexists with acute irrigation water scarcity during spring droughts, while surface storage is severely constrained by high geological leakage rates. To resolve this mismatch, a decentralized WEF architecture is designed that converts surplus renewable electricity into gravity-stored irrigation water, bypassing the need for electrochemical batteries. The supplementary package includes two components: appendices and datasets. The appendices (Appendix A to D) provide the complete mathematical formulation of the bi-objective MILP model, including objective functions for total annualized cost minimization and lifecycle carbon emission calculation, energy and water balance constraints, karst-specific nonlinear leakage functions, geological carrying capacity constraints, and technical operating constraints. Appendix B catalogs all input parameters with their economic, technical, environmental, and geological values and sources. Appendix C presents supplementary results including hourly dispatch schedules and cross-validation against HOMER Pro and GenX. Appendix D provides a complete nomenclature of symbols and abbreviations used in the model. The datasets are provided as Excel files (.xlsx), each containing a cover page with metadata and a data page for direct replication. Five datasets are included: First, input_parameters.xlsx contains 36 key parameters across six categories (meteorological, hydropower, agricultural, geological, economic, technical), corresponding to Table 1 of the paper. Second, optimization_results_pareto.xlsx contains 31 Pareto-optimal solutions from the multi-objective optimization, mapping the trade-off between levelized cost of water and lifecycle carbon emissions, corresponding to Fig 3 and Table 2. Third, scenario_comparison.xlsx provides a three-scenario comparison (diesel baseline, grid-powered, and proposed WEF system) across nine techno-economic and environmental metrics, corresponding to Table 2 and Fig 5. Fourth, time_series_dispatch.xlsx contains 336 hourly dispatch records for wet season (July) and dry season (March) weeks, demonstrating the seasonal mismatch resolution strategy, corresponding to Fig 4 and Appendix C.1. Fifth, sensitivity_analysis.xlsx presents results in both wide and long formats, including sensitivity coefficients, breakeven points, and LCOW responses to ±20% parameter variations, corresponding to Fig 6 and Table C.2. All data are provided under a CC BY-NC 4.0 license. The complete MILP model code in Python/Pyomo is available upon request.